Abstract
In the near future, demand for heterogeneous wireless networking (HWN) is expected to increase. QoS provisioning in these networks is a challenging issue considering the diversity in wireless networking technologies and the existence of mobile users with different communication requirements. In HWNs with their increased complexity, “the curse of dimensionality” problem makes it impractical to directly apply the decision theoretic optimal control methods that are previously used in homogeneous wireless networks to achieve desired QoS levels. In this paper, optimal call admission control policies for HWNs are considered. A decision theoretic framework for the problem is derived by a dynamic programming formulation. We prove that for a two-tier wireless network architecture, the optimal policy has a two-dimensional threshold structure. Further, this structural result is used to design two computationally efficient algorithms, Structured Value Iteration and Structured Update Value Iteration. These algorithms can be used to determine the optimal policy in terms of thresholds. Although the first one is closer in its operation to the conventional Value Iteration algorithm, the second one has a significantly lower complexity. Extensive numerical observations suggest that, for all practical parameter sets, the algorithms always converge to the overall optimal policy. Further, the numerical results show that the proposed algorithms are efficient in terms of time-complexity and in achieving the optimal performance.
Similar content being viewed by others
References
Stemm M, Katz RH (1998) Vertical handoffs in wireless overlay networks. Mob Netw Appl (MONET) 3(4):335–350
Zahran AH, Liang B, Saleh A (2006) Signal threshold adaptation for vertical handoff in heterogeneous wireless networks. ACM/Kluwer Mob Netw Appl (MONET), Special issue on soft radio enabled heterogeneous networks 11(4):625–640
Naghshineh M, Schwartz M (1996) Distributed call admission control in mobile/wireless networks. IEEE JSAC 14(4):711–717
Wu S, Wong K, Li B (2002) A dynamic call admission policy with precision QoS guarantee using stochastic control for mobile wireless networks. IEEE/ACM Trans Netw 10(2):257–271
Ramjee R, Towsley D, Nagarajan R (1997) On optimal call admission control in cellular networks. Wirel Netw 3(1):29–41
Ho C-J, Lea C-T (1999) Improving call admission policies in wireless networks. Wirel Netw 5(4):257–265
Choi J, Kwon T, Choi Y, Naghshineh M (2000) Call admission control for multimedia services in mobile cellular networks: a markov decision approach. In: Fifth IEEE symposium on computers and communications, Proc.ISCC 2000, pp 594–599
Puterman M (1994) Markov decision processes: discrete stochastic dynamic programming. John Wiley & Sons, Inc., New York, NY, USA
Littman M, Dean T, Kaelbling L (1995) On the complexity of solving Markov decision problems. In: Proc. of the eleventh annual conference on uncertainty in artificial intelligence, pp 394–402
Haas Z, Halpern J, Li L, Wicker S (2000) A decision-theoretic approach to resource allocation in wireless multimedia networks. In: Proc. of the 4th ACM international workshop on discrete algorithms and methods for mobile computing and communications, pp 86–95
Altman E, Jimenez T, Koole G (2001) On optimal call admission control in resource-sharing system. IEEE Trans Commun 49(9):1659–1668
Ni J, Tsang DHK, Tatikonda S, Bensaou B (2007) Optimal and structured call admission control policies for resource-sharing systems. IEEE Trans Commun 158–170
Gavious A, Rosberg Z (1994) A restricted complete sharing policy for a stochastic knapsack problem in b-isdn. IEEE Trans Comm 42(7):2375–2379
Ross K, Tsang D (1989) The stochastic knapsack problem. IEEE Trans Commun 37(7):740–747
Nobel R, Tijms H (2000) Optimal control of a queueing system with heterogeneous servers and setup costs. IEEE Trans Automat Contr 45(4):780–784
Varma V, Ramesh S, Wong K, Barton M, Hayward G, Friedhoffer J (2003) Mobility management in integrated umts/wlan networks. In: IEEE Int. Conf. on Comm.(ICC), pp 1048–1053
Fang Y, Chlamtac I, Lin Y-B (1998) Channel occupancy times and handoff rate for mobile computing and pcs networks. IEEE Trans Comput 47(6):679–692
Medepalli K, Gopalakrishnan P, Famolari D, Kodama T (2004) Voice capacity of IEEE 802.11b, 802.11a and 802.11g wireless LANs. In: IEEE Global Telecomm. Conf.(GLOBECOM), pp 1549–1553
Hole D, Tobagi F (2004) Capacity of an ieee 802.11b wireless lan supporting VoIP. IEEE Int Conf Comm 1:196–201
Ross SM (1992) Applied probability models with optimization applications. Dover Publications
Tijms HC (2003) A first course in stochastic models. Wiley
Wood GR (1992) The bisection method in higher dimensions. J Math Program 55:319–337
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was supported in part by a grant from LG Electronics and by Bell Canada through its Bell University Laboratories R&D program.
Appendix
Appendix
Proof of Lemma 1 V k + 1(i,j ) as shown in Eq. 2 consists of nine terms, T 1 ...T 9. Here for clarity, we factor out the common 1/v max term and label the costs by their class number. We define
We use induction on k to prove V k + 1(i,j ) is convex and monotonically non-decreasing in i for every fixed j (proof for convexity in j is similar). This is equal to showing \(\Delta^i V_{k+1}(i+1,\!j\,) \geq \Delta^i V_{k+1}(i,\!j\,)\) and \(\Delta^i V_{k+1}(i,j\,) \geq 0\). The induction hypothesis is the following: V k (i,j ) is (I) monotonically non-decreasing in i and j and (II) convex in i (or j ) for every fixed j (or i ).□
Proof of (I) We show \(\Delta^i V_{k+1}(i,\!j\,) \geq 0\) by evaluating each D m term individually. The basis step is trivial since ∀ (i,j ) ≤ (C c ,C w ): V 0(i,j ) = 0 and ∀ j: V 0(C + 1,j ) = ∞. Consider D 1(i,j ), when λ c is factored out. Let
The same method can be applied directly to T {2,3,5,7}. However, the rest of the terms have to be considered altogether. We first extend D 9(i,j ) as follows,
Although it is straight forward to show that the third and fifth terms above are non-negative, for other terms it is not trivial. We show that every of these terms in D 9(i,j ) if combined with other terms in \(\Delta^i V_{k}(i,\!j\,)\) can be proved to be non-negative. As an example, consider the first term which we can rewrite as
Equation 21 contains η hcc, and from Eq. 2 we see that term T 8 has η hcc as well, so we will use D 8(i,j ). We consider the sum of them:
The other two terms in D 9(i,j ) can be matched with T 4 and T 6 similarly. Hence, \(\Delta^i V_{k+1}(i,\!j\,)= \sum_{m=1}^9 D_m \geq 0\). □
Proof of (II) We need to show \(\Delta^i V_{k+1}(i+1,\!j\,) \geq \Delta^i V_{k+1}(i,\!j\,)\). This is equal to \(\sum_{m=1}^{9} [ D_{m}(i+1,\!j\,)- D_{m}(i,\!j\,)] \geq 0\). Similar Similar to the last part, the basis step is trivial since ∀ (i,j ) ≤ (C c ,C w ): V 0(i,j ) = 0 and ∀ j: V 0(C + 1,j ) = ∞. Let us define Y m (i,j ) = D m (i,j ) − D m (i − 1,j ). Again, we prove terms are non-negative either individually or when combined with other terms. Let us start with Y 1, having factored out λ c :
Again, the proof for Y {2,3,5,7} is the same as for Y 1. For other terms, we have to take an approach similar to the last part. Let us extend Y 9, focusing on the terms containing η hcc:
Separating the terms with η hcc, we obtain
This term has to be evaluated along with Y 8(i + 1,j ) in order to show that the sum of both terms is non-negative. For Y 8(i + 1,j ) we have
and the sum of these two terms is
It is similar to show non-negativity for the other terms of Y m . Since we have ∀ m: Y m ≥ 0, the inequality \(\Delta^i V_{k+1}(i+1,\!j\,) \geq \Delta^i V_{k+1}(i,\!j\,)\) holds, and hence, the cost function is convex. □
Rights and permissions
About this article
Cite this article
Farbod, A., Liang, B. Efficient Structured Policies for Admission Control in Heterogeneous Wireless Networks. Mobile Netw Appl 12, 309–323 (2007). https://doi.org/10.1007/s11036-008-0045-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11036-008-0045-5