Skip to main content
Log in

RPO-MAC: reciprocal Partially observable MAC protocol based on application-value-awareness in VANETs

  • Published:
Wireless Networks Aims and scope Submit manuscript

Abstract

Numerous safety-related applications have rigid requirements for the threshold of latency limitation of message transmissions to enhance safety and traffic efficiency of vehicles on the road in vehicular ad-hoc networks (VANETs). Consequently, a medium access control (MAC) protocol for reliable and fast transmission of safety messages should focus on the maximum transmission delay. Herein, we refer to the concept of packet value and integrate it with the latency of different messages determined by the waiting interval of packets in a queue. Subsequently, we propose a reciprocal revenue function, i.e., the transmission probability of the participants, by considering both its own benefit and the other nodes’ revenue. Additionally, an inter-vehicle cross-layer cooperative game model based on the local optimal utility of participants is constructed. We then theoretically prove the existence of an equilibrium using the partially observable Markov decision process (POMDP), and provide a specific approach for obtaining the channel access probability of a vehicle by using deep reinforcement learning. Finally, the analysis and simulation results in saturated and non-saturated data traffic conditions are presented to evaluate the performance of Reciprocal Partially Observable MAC Protocol (RPO-MAC) proposed in this paper and compare it with the IEEE 802.11p standard protocol. These comparisons demonstrate the advantages of our proposed reciprocal revenue game method in the case of channel network congestion, especially in terms of delay. It is shown that the RPO-MAC protocol can provide a strong support to delay-sensitive safety related applications.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  1. Azarafrooz, M., Chandramouli, R., & Subbalakshmi, K. (2014). Reciprocity, fairness and learning in medium access control games. Computer Communication, 46, 22–28.

    Article  Google Scholar 

  2. Babu, S., Patra, M., & Murthy, C. S. R. (2015). A novel context-aware variable interval mac protocol to enhance event-driven message delivery in ieee 802.11 p/wave vehicular networks. Vehicular Communications, 2(3), 172–183.

    Article  Google Scholar 

  3. Burton, M., & Hill, G. (2009). 802.11 arbitration. White Paper. Durham: Certified Wireless Network Professional Inc.

    Google Scholar 

  4. Chen, L., Low, S. H., & Doyle, J. C. (2010). Random access game and medium access control design. IEEE/ACM Transactions on Networking (TON), 18(4), 1303–1316.

    Article  Google Scholar 

  5. Cheng, N., Zhang, N., Lu, N., Shen, X., Mark, J. W., & Liu, F. (2013). Opportunistic spectrum access for cr-vanets: A game-theoretic approach. IEEE Transactions on Vehicular Technology, 63(1), 237–251.

    Article  Google Scholar 

  6. Committee ICSLMS, et al. (1999) Wireless lan medium access control (mac) and physical layer (phy) specifications. ANSI/IEEE Std 80211-1999

  7. Cui, T., Chen, L., & Low, S. H. (2008). A game-theoretic framework for medium access control. IEEE Journal on Selected Areas in Communications, 26(7), 1116–1127.

    Article  Google Scholar 

  8. Dang, D. N. M., Hong, C. S., Lee, S., & Huh, E. N. (2014). An efficient and reliable mac in vanets. IEEE Communications Letters, 18(4), 616–619.

    Article  Google Scholar 

  9. Fazio, P., De Rango, F., & Sottile, C. (2016). A predictive cross-layered interference management in a multichannel mac with reactive routing in vanet. IEEE Transactions on Mobile Computing, 15(8), 1850–1862.

    Article  Google Scholar 

  10. Hajek, B. (1984). Optimal control of two interacting service stations. IEEE transactions on automatic control, 29(6), 491–499.

    Article  MathSciNet  Google Scholar 

  11. Hegde, N., & Proutiere, A. (2012). Simulation-based optimization algorithms with applications to dynamic spectrum access. In: Information Sciences and Systems (CISS), 2012 46th Annual Conference on, IEEE, pp 1–6

  12. Jiang, L., Shah, D., Shin, J., & Walrand, J. (2010). Distributed random access algorithm: Scheduling and congestion control. IEEE Transactions on Information Theory, 56(12), 6182–6207.

    Article  MathSciNet  Google Scholar 

  13. Ju, P., & Song, W. (2016). Repeated game analysis for cooperative mac with incentive design for wireless networks. IEEE Transactions on Vehicular Technology, 65(7), 5045–5059.

    Article  Google Scholar 

  14. Karkus, P., Hsu, D., & Lee, W.S. (2017). Qmdp-net: Deep learning for planning under partial observability. In: Advances in Neural Information Processing Systems, pp 4694–4704

  15. Kenney, J. B. (2011). Dedicated short-range communications (dsrc) standards in the united states. Proceedings of the IEEE, 99(7), 1162–1182.

    Article  Google Scholar 

  16. Khanduri, R., & Rattan, S. (2013). Performance comparison analysis between ieee 802.11 a/b/g/n standards. International Journal of Computer Applications, 78(1), 13–20.

    Article  Google Scholar 

  17. Kim, T. H., Ni, J., Srikant, R., & Vaidya, N. H. (2013). Throughput-optimal csma with imperfect carrier sensing. IEEE/ACM Transactions on Networking (TON), 21(5), 1636–1650.

    Article  Google Scholar 

  18. Kumar, P. R., & Varaiya, P. (2015). Stochastic systems: Estimation, identification, and adaptive control (Vol. 75). Philadelphia: SIAM.

    Book  Google Scholar 

  19. Kwon, Y., & Rhee, B. (2016). Bayesian game-theoretic approach based on 802.11 p mac protocol to alleviate beacon collision under urban vanets. International Journal of Automotive Technology, 17(1), 183–191.

    Article  Google Scholar 

  20. Lee, J. H., Ernst, T., & Ma, X. (2014). Performance analysis of secure beaconing messages for geonetworking. Security and Communication Networks, 7(12), 2555–2563.

    Article  Google Scholar 

  21. Lee, J.-W., Chiang, M., & Calderbank, A. R. (2006). Utility-optimal medium access control: Reverse and forward engineering. In Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications (pp. 1–13). Barcelona, Spain. https://doi.org/10.1109/INFOCOM.2006.252.

  22. Lee, J. W., Tang, A., Huang, J., Chiang, M., & Calderbank, A. R. (2007). Reverse-engineering mac: A non-cooperative game model. IEEE Journal on Selected Areas in Communications, 25(6), 1135–1147.

    Article  Google Scholar 

  23. Liu, L., Xia, W., & Shen, L. (2009). An adaptive multi-channel mac protocol with dynamic interval division in vehicular environment. In: 2009 First International Conference on Information Science and Engineering, IEEE, pp 2534–2537

  24. Maalej, Y., Abderrahim, A., Guizani, M., Hamdaoui, B., & Balti, E. (2016). Advanced activity-aware multi-channel operations1609. 4 in vanets for vehicular clouds. In: 2016 IEEE Global Communications Conference (GLOBECOM), IEEE, pp 1–6

  25. MacDermed, L., Isbell, C., & Weiss, L. (2011). Markov games of incomplete information for multi-agent reinforcement learning. In: Workshops at the Twenty-Fifth AAAI Conference on Artificial Intelligence

  26. Nayyar, A., Mahajan, A., & Teneketzis, D. (2011). Optimal control strategies in delayed sharing information structures. IEEE Transactions on Automatic Control, 56(7), 1606–1620.

    Article  MathSciNet  Google Scholar 

  27. Nguyen, V., Kim, O. T. T., Pham, C., Oo, T. Z., Tran, N. H., Hong, C. S., & Huh, E. N. (2018). A survey on adaptive multi-channel mac protocols in vanets using markov models. IEEE Access, 6, 16493–16514.

    Article  Google Scholar 

  28. Ni, J., Tan, B., & Srikant, R. (2012). Q-csma: Queue-length-based csma/ca algorithms for achieving maximum throughput and low delay in wireless networks. IEEE/ACM Transactions on Networking (TON), 20(3), 825–836.

    Article  Google Scholar 

  29. Omar, H. A., Zhuang, W., & Li, L. (2012). Vemac: A tdma-based mac protocol for reliable broadcast in vanets. IEEE transactions on mobile computing, 12(9), 1724–1736.

    Article  Google Scholar 

  30. Qiu, H. J., Ho, I. W. H., Chi, K. T., & Xie, Y. (2015). A methodology for studying 802.11 p vanet broadcasting performance with practical vehicle distribution. IEEE transactions on vehicular technology, 64(10), 4756–4769.

    Article  Google Scholar 

  31. Rajagopalan, S., Shah, D., & Shin, J. (2009). Network adiabatic theorem: an efficient randomized protocol for contention resolution. In: Proceedings of the eleventh international joint conference on Measurement and modeling of computer systems, pp 133–144

  32. Reeves, D.M., & Wellman, M.P. (2004). Computing best-response strategies in infinite games of incomplete information. In: Proceedings of the 20th conference on Uncertainty in artificial intelligence, AUAI Press, pp 470–478

  33. Sahoo, J., Wu, E. H. K., Sahu, P. K., & Gerla, M. (2013). Congestion-controlled-coordinator-based mac for safety-critical message transmission in vanets. IEEE Transactions on Intelligent Transportation Systems, 14(3), 1423–1437.

    Article  Google Scholar 

  34. Smith, J. E., & McCardle, K. F. (2002). Structural properties of stochastic dynamic programs. Operations Research, 50(5), 796–809.

    Article  MathSciNet  Google Scholar 

  35. Song, C., Tan, G., Yu, C., Ding, N., & Zhang, F. (2017). Apdm: An adaptive multi-priority distributed multichannel mac protocol for vehicular ad hoc networks in unsaturated conditions. Computer Communications, 104, 119–133.

    Article  Google Scholar 

  36. Vasal, D., & Anastasopoulos, A. (2014). Stochastic control of relay channels with cooperative and strategic users. IEEE Transactions on Communications, 62(10), 3434–3446.

    Article  Google Scholar 

  37. Wang, J., Lang, P., Zhu, J., Deng, W., & Xu, S. (2018). Application-value-awareness cross-layer mac cooperative game for vehicular networks. Vehicular Communications, 13, 27–37.

    Article  Google Scholar 

  38. Wang, Q., Leng, S., Fu, H., & Zhang, Y. (2011a). An ieee 802.11 p-based multichannel mac scheme with channel coordination for vehicular ad hoc networks. IEEE transactions on intelligent transportation systems, 13(2), 449–458.

    Article  Google Scholar 

  39. Wang, Q., Leng, S., Zhang, Y., & Fu, H. (2011b). A qos supported multi-channel mac for vehicular ad hoc networks. In: 2011 IEEE 73rd Vehicular Technology Conference (VTC Spring), IEEE, pp 1–5

  40. Yagan, D., & Tham, C. K., (2007). Coordinated reinforcement learning for decentralized optimal control. In: Approximate Dynamic Programming and Reinforcement Learning 2007. ADPRL 2007. IEEE International Symposium on, IEEE, pp. 296–302

  41. Yao, Y., Zhang, K., & Zhou, X. (2017). A flexible multi-channel coordination mac protocol for vehicular ad hoc networks. IEEE Communications Letters, 21(6), 1305–1308.

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by National Nature Science Foundation[61572229 and 6171101066]; Jilin Provincial Science and Technology Development Foundation [20170204074GX and 20180201068GX]; Jilin Provincial International Cooperation Foundation [20180414015GH]; CERNET Innovation Project [NGII20170413].

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Xuejie Liu.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix A

Proof

When \(t=1\) , it can be assumed that the initial channel usage state is idle, and obviously all primitive values \(R_i^1\), \(R_j^i\) of any two vehicles are independent of each other.

Suppose that when \(t=k,k \in N^+\), the conclusion is true, i.e.,

$$\begin{aligned} {\mathbb {P}}^{\varvec{\psi }} \left( \varvec{r}^{\varvec{1:k}}|o^{1:k-1}\right) =\prod _{i=1}^n{\mathbb {P}}^{\phi _i}\left( r_i^{1:k}|o^{1:k-1}\right) . \end{aligned}$$

Then, when \(t=k+1\),

$$\begin{aligned}&{\mathbb {P}}^{\varvec{\psi }} \left( \varvec{r^{1:k+1}},o^{1:k}\right) \\&\quad ={\mathbb {P}}\left( {o}^{1:k-1}\right) {\mathbb {P}}\left( \varvec{r}^{\varvec{1:k}}|{o}^{1:k-1}\right) {\mathbb {P}}\left( {o}^k|\varvec{r}^{\varvec{1:k}},{o}^{1:k-1}\right) *\\&\qquad {\mathbb {P}}^{\varvec{\psi }}\left( \varvec{r^{k+1}}|\varvec{r}^{\varvec{1:k}},{o}^{1:k}\right) \\&={\mathbb {P}}\left( {o}^{1:k-1}\right) {\mathbb {P}}\left( \varvec{r}^{\varvec{1:k}}|{o}^{1:k-1}\right) {\mathbb {P}}^{\varvec{\psi }}\left( {o}^{1:k}|\varvec{a^k}\right) \\&\qquad {\mathbb {P}}^{\varvec{\psi }}\left( \varvec{r^{k+1}}|\varvec{a^k},\varvec{r^k},{o}^{1:k}\right) \\&={\mathbb {P}}\left( {o}^{1:k-1}\right) \prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k-1}\right) *{\mathbb {P}}\left( {o}^{1:k}|\varvec{r}^{\varvec{1:k}},{o}^{1:k-1}\right) \\&\qquad \prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^k,{o}^{1:k}\right) \\&{\mathbb {P}}\left( r_i^{1:k+1},{o}^{1:k}\right) ={\mathbb {P}}\left( {o}^{1:k-1}\right) {\mathbb {P}}\left( r_i^{1:k}|{o}^{1:k-1}\right) \\&\qquad {\mathbb {P}}\left( {o}^{1:k}|r_i^{1:k},{o}^{1:k-1}\right) {\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^{1:k},{o}^{1:k}\right) \\&{\mathbb {P}}\left( {o}^{1:k}\right) ={\mathbb {P}}\left( {o}^{1:k-1}\right) \ {\mathbb {P}}\left( {o}^{1:k}|{o}^{1:k-1}\right) \\ \end{aligned}$$

For any vehicle, i

$$\begin{aligned}&{\mathbb {P}}^\psi \left( r_i^{1:k+1}| o^{1:k}\right) \\&\quad =\dfrac{{\mathbb {P}}(r_i^{1:k+1},o^{1:k})}{{\mathbb {P}}(o^{1:k})}\\&\quad ={\mathbb {P}}\left( r_i^{1:k}|{o}^{1:k-1}\right) {\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^{1:k},{o}^{1:k}\right) \\&\qquad \dfrac{{\mathbb {P}}\left( {o}^{1:k}|r_i^{1:k},{o}^{1:k-1}\right) }{{\mathbb {P}}\left( {o}^{1:k}|{o}^{1:k-1}\right) }\\&\quad ={\mathbb {P}}\left( r_i^{1:k}|{o}^{1:k-1}\right) {\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^{1:k},{o}^{1:k}\right) \dfrac{{\mathbb {P}}\left( {o}^{1:k},r_i^{1:k}\right) }{{\mathbb {P}}\left( r_i^{1:k},{o}^{1:k-1}\right) }\\&\qquad \dfrac{{\mathbb {P}}({o}^{1:k-1})}{{\mathbb {P}}({o}^{1:k})}\\&\quad ={\mathbb {P}}\left( r_i^{1:k}|{o}^{1:k-1}\right) {\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^{1:k},{o}^{1:k}\right) \dfrac{{\mathbb {P}}\left( r_i^{1:k}|{o}^{1:k}\right) }{{\mathbb {P}}\left( r_i^{1:k}|{o}^{1:k-1}\right) }\\&\quad ={\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^{1:k},{o}^{1:k}\right) {\mathbb {P}}\left( r_i^{1:k}|{o}^{1:k}\right) \\ \end{aligned}$$

So, when \(t=k+1\),

$$\begin{aligned}&{\mathbb {P}}^{\varvec{\psi }} \left( \varvec{r^{1:k+1}}|o^{1:k}\right) \\&\quad =\dfrac{{\mathbb {P}}^{\varvec{\psi }} \left( \varvec{r^{1:k+1}},o^{1:k}\right) }{{\mathbb {P}}(o^{1:k})}\\&\quad =\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k-1}\right) \prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^k,{o}^{1:k}\right) *\\&\qquad \dfrac{{\mathbb {P}}^{\varvec{\psi }}\left( {o}^{1:k}|\varvec{r}^{\varvec{1:k}},{o}^{1:k-1}\right) }{{\mathbb {P}}\left( o^{1:k}|{o}^{1:k-1}\right) }\\&\quad =\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k-1}\right) \prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^k,{o}^{1:k}\right) *\\&\qquad \dfrac{{\mathbb {P}}^{\varvec{\psi }}\left( {o}^{1:k},\varvec{r}^{\varvec{1:k}}\right) }{{\mathbb {P}}^{\varvec{\psi }}\left( \varvec{r}^{\varvec{1:k}},{o}^{1:k-1}\right) }\frac{{\mathbb {P}}({o}^{1:k-1})}{{\mathbb {P}}({o}^{1:k})}\\&\quad =\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k-1}\right) \prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^k,{o}^{1:k}\right) *\\&\qquad \dfrac{{\mathbb {P}}^{\varvec{\psi }}\left( \varvec{r}^{\varvec{1:k}}|{o}^{1:k}\right) }{{\mathbb {P}}^{\varvec{\psi }}\left( \varvec{r}^{\varvec{1:k}}|{o}^{1:k-1}\right) }\\&\quad =\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k-1}\right) \prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^k,{o}^{1:k}\right) *\\&\qquad \dfrac{\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k}\right) }{\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k-1}\right) }\\&\quad =\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{k+1}|r_i^k,{o}^{1:k}\right) \prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k}|{o}^{1:k}\right) \\&\quad =\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:k+1}|{o}^{1:k}\right) \\ \end{aligned}$$

Thus,

$$\begin{aligned} {\mathbb {P}}^{\varvec{\psi }} \left( \varvec{r^{1:t}}|o^{1:t-1}\right) =\prod _{i=1}^n{\mathbb {P}}^{\psi _i}\left( r_i^{1:t}|o^{1:t-1}\right) \end{aligned}$$

\(\square\)

Appendix B

Proof

$$\begin{aligned}&{\mathbb {P}}^{\varvec{\psi _{-i}}} \left( r_i^{t+1},o^{1:t}|r_i^{1:t},o^{1:t-1},a_i^{1:t}\right) \\&\quad ={\mathbb {P}}^{\varvec{\psi _{-i}}}\left( r_i^{t+1}|r_i^{1:t},o^{1:t},a_i^{1:t}\right) *{\mathbb {P}}^{\varvec{\psi _{-i}}} \left( o^{1:t}|r_i^{1:t},o^{1:t-1},a_i^{1:t}\right) \\&\quad ={\mathbb {P}}^{\varvec{\psi _{-i}}}\left( r_i^{t+1}|r_i^{t},o^{t},a_i^{t}\right) *{\mathbb {P}}^{\varvec{\psi _{-i}}} \left( \varvec{a_{-i}^t}|r_i^{1:t},o^{1:t-1},a_i^{1:t}\right) \\&\quad ={\mathbb {P}}^{\varvec{\psi _{-i}}}\left( r_i^{t+1}|r_i^{t},o^{t},a_i^{t}\right) *{\mathbb {P}}^{\varvec{\psi _{-i}}} \left( \varvec{a_{-i}^t}|o^{1:t-1}\right) \\&\quad ={\mathbb {P}}^{\varvec{\psi _{-i}}} \left( r_i^{t+1},o^{1:t}|r_i^{t},o^{1:t-1},a_i^{t}\right) \end{aligned}$$

where \(a_i^t\) is a function of \(r_i^t\). \({\mathbb {P}}^{\varvec{\psi _{-i}}}\left( \varvec{a_{-i}^t}|r_i^{1:t},o^{1:t-1},a_i^{1:t}\right)\) can be switched into \({\mathbb {P}}^{\varvec{\psi _{-i}}} \left( \varvec{a_{-i}^t}|o^{1:t-1}\right)\) because each action \(a_i\) is conditionally independent given the historical observations \(o^{1:t-1}\)(Lemma 1).

For the second part,

$$\begin{aligned}&{\mathbb {E}}^{\varvec{\psi _{-i}}} \left\{ u_i\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}\right) |r_i^{1:t},o^{1:t-1},a_i^{1:t}\right\} \\&\quad =\sum _{\varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}}u_i\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}\right) {\mathbb {P}}^{\varvec{\psi _{-i}}}\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}|r_i^{1:t},o^{1:t-1},a_i^{1:t}\right) \\&\quad =\sum _{\varvec{r_{-i}^t,a_{-i}^t}}u_i\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}\right) {\mathbb {P}}^{\varvec{\psi _{-i}}}\left( \varvec{r_{-i}^t,a_{-i}^t}|r_i^{1:t},o^{1:t-1},a_i^{1:t}\right) \\&\quad =\sum _{\varvec{r_{-i}^t,a_{-i}^t}}u_i\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}\right) {\mathbb {P}}^{\varvec{\psi _{-i}}}\left( \varvec{r_{-i}^t,a_{-i}^t}|o^{1:t-1}\right) \\&\quad ={\mathbb {E}}^{\varvec{\psi _{-i}}} \left\{ u_i\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}\right) |r_i^{t},o^{1:t-1},a_i^{t}\right\} \\&\quad =\hat{u_i}\left( r_i^{t},o^{1:t-1},a_i^{t}\right) \end{aligned}$$

\(\square\)

Appendix C

Proof

Fix \(\varvec{\phi }\),

$$\begin{aligned}&b_i^{t+1}\left( r_i^{t+1}\right) ={\mathbb {P}} \left( r_i^{t+1}|o^{1:t},\varvec{\gamma }^{\varvec{1:t}}\right) \\&\quad =\sum _{\varvec{r}^{\varvec{t}}}{\mathbb {P}} \left( r_i^{t+1},\varvec{r}^{\varvec{t}}|o^{1:t},\varvec{\gamma }^{\varvec{1:t}}\right) \\&\quad =\sum _{\varvec{r}^{\varvec{t}}}{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{r}^{\varvec{t}}|o^{1:t},\varvec{\gamma }^{\varvec{1:t}}\right) {\mathbb {P}}^{\varvec{\psi }}\left( r_i^{t+1}|r_i^t,o^{1:t},\varvec{\gamma }^{\varvec{1:t}}\right) \\&\quad =\sum _{\varvec{r}^{\varvec{t}}}{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{r}^{\varvec{t}}|o^{1:t},\varvec{\gamma }^{\varvec{1:t}}\right) {\mathbb {P}}\left( r_i^{t+1}|o^t,\gamma _i^t(r_i^t)\right) \end{aligned}$$

which

$$\begin{aligned}&{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{r}^{\varvec{t}}|o^{1:t},\varvec{\gamma }^{\varvec{1:t}}\right) \\&\quad =\dfrac{{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{r}^{\varvec{t}},o^t|o^{1:t-1}, \varvec{\gamma }^{\varvec{1:t}}\right) }{\sum _{\varvec{{\hat{r}}^t}}{\mathbb {P}}^{\varvec{\phi }} \left( \varvec{{\hat{r}}^t},o^t|o^{1:t-1},\varvec{\gamma }^{\varvec{1:t}}\right) }\\&\quad =\dfrac{{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{r}^{\varvec{t}}|o^{1:t-1}, \varvec{\gamma }^{\varvec{1:t}}\right) {\mathbb {P}}^{\varvec{\phi }} \left( o^t|o^{1:t-1},\varvec{\gamma }^{\varvec{1:t}},\varvec{r}^{\varvec{t}}\right) }{\sum _{\varvec{{\hat{r}}^t}}{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{{\hat{r}}^t}|o^{1:t-1}, \varvec{\gamma }^{\varvec{1:t}}\right) {\mathbb {P}}^{\varvec{\phi }}\left( o^t|o^{1:t-1}, \varvec{\gamma }^{\varvec{1:t}},\varvec{{\hat{r}}^t}\right) }\\&\quad =\dfrac{{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{r}^{\varvec{t}}|o^{1:t-1},\varvec{\gamma }^{\varvec{1:t}}\right) {\mathbb {P}}^{\varvec{\phi }}\left( o^t|\varvec{\gamma ^{t}(r^t)}\right) }{\sum _{\varvec{{\hat{r}}^t}}{\mathbb {P}}^{\varvec{\phi }}\left( \varvec{{\hat{r}}^t}|o^{1:t-1},\varvec{\gamma }^{\varvec{1:t}}\right) {\mathbb {P}}^{\varvec{\phi }}\left( o^t|\varvec{\gamma ^{t}({\hat{r}}^t)}\right) } \end{aligned}$$

Since packet values of all vehicles \(R_i^{1:t}\) are conditionally independent with the observations \(o^{1:t-1}\) (Lemma 1), we can get:

$$\begin{aligned} {\mathbb {P}}^{\varvec{\phi }}(\varvec{r}^{\varvec{t}}|{o}^{1:t},\varvec{\gamma }^{\varvec{1:t}})=\dfrac{\prod _{i=1}^{N}{b_i^t(r_i^t)}{\mathbb {P}}^{\varvec{\phi }}\left( {o}^t|\varvec{\gamma }^{\varvec{t}}(\varvec{r}^{\varvec{t}})\right) }{\sum _{\varvec{{\hat{r}}^t}}{\prod _{i=1}^{N}{b_i^t({{\hat{r}}}_i^t)}{\mathbb {P}}^{\varvec{\phi }}({o}^t|\varvec{\gamma }^{\varvec{t}}({\varvec{{\hat{r}}^t}}))}} \end{aligned}$$

So, \(b_i^{t+1}\) can be expressed as follows:

$$\begin{aligned}&b_i^{t+1}\left( r_i^{t+1}\right) \\&\quad =\sum _{\varvec{r}^{\varvec{t}}}{\mathbb {P}}\left( r_i^{t+1}|o^t,\gamma _i^t(r_i^t)\right) \dfrac{\prod _{i=1}^{N}{b_i^t(r_i^t)}{\mathbb {P}}^{\varvec{\phi }}\left( {o}^t|\varvec{\gamma }^{\varvec{t}}(\varvec{r}^{\varvec{t}})\right) }{\sum _{\varvec{{\hat{r}}^t}}{\prod _{i=1}^{N}{b_i^t({{\hat{r}}}_i^t)}{\mathbb {P}}^{\varvec{\phi }}({o}^t|\varvec{\gamma }^{\varvec{t}}({\varvec{{\hat{r}}^t}}))}} \end{aligned}$$

We can further simplify this function because the delay of vehicle i in time \(t + 1\) is meaningful only when it chooses to wait at time t.Thus,

$$\begin{aligned}&b_i^{t+1}\left( r_i^{t+1}\right) \\&\quad =\sum _{\varvec{r}^{\varvec{t}}}\varvec{1}_{\left\{ W\right\} }({\gamma _i^t(r_i^t)}){\mathbb {P}}\left( r_i^{t+1}|A_i = W,r_i^t,{o}^t\right) \\&\qquad \dfrac{\prod _{i=1}^{N}{b_i^t(r_i^t)}{\mathbb {P}}^{\varvec{\phi }}\left( {o}^t|A_i = W,\varvec{\gamma _{-i}^t}(\varvec{r_{-i}^t})\right) }{\sum _{\varvec{{\hat{r}}^t}}{\prod _{i=1}^{N}{b_i^t({{\hat{r}}}_i^t)}{\mathbb {P}}^{\varvec{\phi }}({o}^t|A_i = W,\varvec{\gamma _{-i}^t}({\varvec{{\hat{r}}^t_{-i}}}))}}\\&\quad =\dfrac{\sum _{r_i^t}\varvec{1}_{\left\{ W\right\} }({\gamma _i^t(r_i^t)}){\mathbb {P}}\left( r_i^{t+1}|A_i=W,r_i^t,{o}^t\right) b_i^t(r_i^t)\sum _{\varvec{r_{-i}^t}}\prod _{j=1,j\not =i}^{N}b_j^t(r_j^t){\mathbb {P}}\left( {o}^t|\varvec{\gamma _{-i}^t}(\varvec{r_{-i}^t})\right) }{\sum _{{\hat{r}}_i^t}b_i^t({\hat{r}}_i^t)\sum _{\varvec{{\hat{r}}_{-i}^t}}\prod _{j=1,j\not =i}^{N}{b_j^t({{\hat{r}}}_j^t)}{\mathbb {P}}({o}^t|\varvec{\gamma _{-i}^t}({\varvec{{\hat{r}}^t_{-i}}}))}\\&\quad =\dfrac{\sum _{r_i^t}\varvec{1}_{\left\{ W\right\} }({\gamma _i^t(r_i^t)}){\mathbb {P}}\left( r_i^{t+1}|A_i=W,r_i^t,{o}^t\right) b_i^t(r_i^t)}{\sum _{{\hat{r}}_i^t}b_i^t({\hat{r}}_i^t)}\\&\quad =H_i\left( b_i^t,\gamma _i^t,o^t\right) \end{aligned}$$

where \(\varvec{1}(\cdot )\) is the indicator function and H is independent of policy \(\psi\). \(\square\)

Appendix D

Proof

$$\begin{aligned}&{\mathbb {P}}\left( \varvec{b^{t+1}}|\varvec{b^{1:t}},\varvec{\gamma }^{\varvec{1:t}}\right) \\&\quad =\sum _{{o}^t}{\mathbb {P}}\left( \varvec{b}^{\varvec{t}+{\mathbf {1}}},{o}^t|\varvec{b}^{{\mathbf {1}}:\varvec{t}},\varvec{\gamma }^{{\mathbf {1}}:\varvec{t}}\right) \\&\quad =\sum _{{o}^t}\varvec{1}_{\left\{ H\left( \varvec{b}^{\varvec{t}},\varvec{\gamma }^{\varvec{t}},{o}^t\right) \right\} }(\varvec{b^{t+1}}){\mathbb {P}}\left( {o}^t|\varvec{b}^{{\mathbf {1}}:\varvec{t}},\varvec{\gamma }^{{\mathbf {1}}:\varvec{t}}\right) \\&\quad =\sum _{{o}^t,\varvec{r}^{\varvec{t}}}\varvec{1}_{\left\{ H\left( \varvec{b}^{\varvec{t}},\varvec{\gamma }^{\varvec{t}},{o}^t\right) \right\} }(\varvec{b^{t+1}}){\mathbb {P}}\left( {o}^t|\varvec{\gamma ^t(r^t)}\right) {\mathbb {P}}\left( \varvec{r}^{\varvec{t}}|\varvec{b}^{{\mathbf {1}}:\varvec{t}},\varvec{\gamma }^{{\mathbf {1}}:\varvec{t}}\right) \\&\quad =\sum _{{o}^t,\varvec{r}^{\varvec{t}}}\varvec{1}_{\left\{ H\left( \varvec{b}^{\varvec{t}},\varvec{\gamma }^{\varvec{t}},{o}^t\right) \right\} }(\varvec{b^{t+1}}){\mathbb {P}}\left( {o}^t|\varvec{\gamma ^t(r^t)}\right) \prod _{i=1}^{N}{b_i(r_i^t)}\\&\quad ={\mathbb {P}}\left( \varvec{b^{t+1}}|\varvec{b^{t}},\varvec{\gamma }^{\varvec{t}}\right) \\&{\mathbb {E}}\left\{ u_i\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}\right) |\varvec{b^{1:t}},\varvec{\gamma }^{\varvec{1:t}}\right\} \\&\quad =\sum _{\varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}}u_i(\varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}){\mathbb {P}}\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}|\varvec{b^{1:t},\gamma ^{1:t}}\right) \\&\quad =\sum _{\varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}}u_i(\varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}})\varvec{1}_{\left\{ \varvec{\gamma ^t(r^t)}\right\} }(\varvec{a^t}){\mathbb {P}}\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}|\varvec{b^{1:t},\gamma ^{1:t}}\right) \\&\quad =\sum _{\varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}}u_i(\varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}})\varvec{1}_{\left\{ \varvec{\gamma ^t(r^t)}\right\} }(\varvec{a^t})\prod _{i=1}^{N}{b_i(r_i^t)}\\&\quad ={\mathbb {E}} \left\{ u_i\left( \varvec{r}^{\varvec{t}},\varvec{a}^{\varvec{t}}\right) |\varvec{b^{t}},\varvec{\gamma }^{\varvec{t}}\right\} \\&\quad =\hat{u_i}\left( \varvec{b^{t}},\varvec{\gamma }^{\varvec{t}}\right) \\ \end{aligned}$$

\(\square\)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, J., Guo, X., Liu, X. et al. RPO-MAC: reciprocal Partially observable MAC protocol based on application-value-awareness in VANETs. Wireless Netw 27, 2509–2528 (2021). https://doi.org/10.1007/s11276-021-02602-5

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11276-021-02602-5

Keywords

Navigation