Results 1 
8 of
8
Approximate Bayesian inference by adaptive quantization of the hypothesis space
 in Proc. 25th Int. Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering
, 2005
"... Abstract. We introduce a method for making approximate Bayesian inference based on quantizing the hypothesis space and repartitioning it as observations become available. The method relies on approximating an optimal inference by using a probability distribution for quantized intervals of the unknow ..."
Abstract

Cited by 3 (2 self)
 Add to MetaCart
(Show Context)
Abstract. We introduce a method for making approximate Bayesian inference based on quantizing the hypothesis space and repartitioning it as observations become available. The method relies on approximating an optimal inference by using a probability distribution for quantized intervals of the unknown quantity, and by adjusting the intervals so as to obtain higher resolution in regions of higher probability, and vice versa. We repartition the hypothesis space adaptively with the aim of maximizing the mutual information between the approximate distribution and the exact distribution. It is shown that this approach is equivalent to maximizing the entropy of the approximate distribution, and we provide lowcomplexity algorithms for approximating multidimensional posterior distributions with tunable complexity/performance. The resulting quantized distribution for a onedimensional case can be visualized as a histogram where each bar has equal area, but in general unequal width. The method can be used to provide adaptive quantization of arbitrary data sequences, or to approximate the posterior expectation of for instance some loss function by summing over a prespecified number of terms.
C.Levcopoulos and O.Petersson. Adaptive Heapsort
 Journal of Alg
, 1993
"... Abstract—It is possible to improve the spectral efficiency of cellular systems by dynamically repartitioning the available bandwidth between different interfering and noninterfering subareas within and among sectors. Here, we investigate the problem of maximizing the expected system throughput in a ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
(Show Context)
Abstract—It is possible to improve the spectral efficiency of cellular systems by dynamically repartitioning the available bandwidth between different interfering and noninterfering subareas within and among sectors. Here, we investigate the problem of maximizing the expected system throughput in a twosector area by dynamic bandwidth partitioning between two transmission modes: 1) reusing bandwidth in the lowinterference area near either of the base stations and 2) using bandwidth for macrodiversity or singlebase station transmission to avoid interference. Mode 2 is typically useful for giving users near the cell border higher bit rates. The suggested solution adapts the bandwidth partitioning to reflect local transmission capacities and bandwidth demands. Thus, it automatically decides on whether both transmission modes should be used and how much bandwidth should be used in each mode. The solution requires only limited knowledge of the future arrival rates and channel qualities and uses probability theory to find a robust bandwidth partitioning. Finally, we discuss access control and when to switch a user between the transmission modes to achieve high spectral efficiency and some minimum average quality of service. Index Terms—Diversity methods, land mobile radio cellular systems, scheduling. I.
Priority Tasks Allocation through the Maximum Entropy Principle
 Proceedings of the 8th WSEAS International Conference on Automation and Information
"... Abstract: Priority scheduling assigns priorities via some task policy. In this work we introduce a priority tasks allocation model for softrealtime systems. Task entities are analyzed as continuous stochastic processes with time restrictions specified by probability density functions with stationa ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Abstract: Priority scheduling assigns priorities via some task policy. In this work we introduce a priority tasks allocation model for softrealtime systems. Task entities are analyzed as continuous stochastic processes with time restrictions specified by probability density functions with stationary parameters over long periods of time. We present a technique for task allocation based on the maximum entropy principle as an element of tasks differentiation. The entropy measurement is proposed for hierarchy task allocation.
SpaceTime Power Scheduling of MIMO Links—Fairness and QoS Considerations
"... Abstract—Power control, beamforming and link scheduling are all important operations to improve the powerandspectral efficiency of networks of multipleinput multipleoutput (MIMO) wireless links. We call a joint optimization of the above operations the spacetime power scheduling (STPS) scheme. ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
(Show Context)
Abstract—Power control, beamforming and link scheduling are all important operations to improve the powerandspectral efficiency of networks of multipleinput multipleoutput (MIMO) wireless links. We call a joint optimization of the above operations the spacetime power scheduling (STPS) scheme. The STPS scheme is formulated as joint optimization of the transmitter covariance matrices of all active MIMO links over all dimensions of space and time, which includes the dimension of frequency as a dual form of time. In this paper, we address the proportional fair (PF) and qualityofservice (QoS) issues of the STPS scheme, which are important for networks with asymmetric topology and/or asymmetric traffic demands. Both slow fading channels and fast fading channels are considered. We demonstrate that the PFSTPS scheme provides a very attractive tradeoff between sum capacity and rate distribution for asymmetric links. We also demonstrate that the QoSSTPS scheme has a much higher powerandspectral efficiency than the previously existing QoS based scheme that do not exploit the temporal freedom. Efficient optimization algorithms for both PFSTPS and QoSSTPS are provided. The STPS scheme is a centralized cooperative scheme which requires a scheduler. For ad hoc networks, this scheduler can be elected adaptively among eligible nodes in the network. Index Terms—MIMO systems, proportional fairness, qualityofservice, space–time power scheduler. I.
1 On Scheduling and Adaptive Modulation with Limited Channel Feedback
"... The throughput gained by employing adaptive modulation and scheduling in cellular communications systems depend on the amount of channel feedback used and the quality of the quantization of the channel information. We discuss the implications of using a rate quantization which maximizes the expected ..."
Abstract
 Add to MetaCart
(Show Context)
The throughput gained by employing adaptive modulation and scheduling in cellular communications systems depend on the amount of channel feedback used and the quality of the quantization of the channel information. We discuss the implications of using a rate quantization which maximizes the expected downlink throughput, and give a procedure for choosing the optimal number of bits for feedback. It turns out that unless the number of users is much less than the number of symbols that constitute a time slot, no more than 1 bit should be used for feedback. We also note that utilizing multiuser diversity, i.e. scheduling transmissions to increase throughput, has a theoretical performance advantage as compared to fixed channel access when the amount of channel information is reduced. On the other hand, unfairness increases, and due to an inherent sensitivity to incorrectly chosen quantization levels, there is a risk of occasional drastic performance drops, to levels below that of fixed access. We further investigate online adaptation of quantization levels as the channels and the number of users vary. A scheduling and adaptive quantization policy for high throughput and qualityofservice requirements in terms of hard interaccess bounds is developed which is shown to combine the advantages of multiuser diversity without compromising fairness. The policy is wellsuited for the multiplechannel case, gives a considerable throughput gain with little channel feedback, and facilitates realtime services.
Summary
"... Ericsson in Göteborg. Uppsala University acted as lead partner. The objectives were to develop and to study algorithms and system design for future wireless packet data transmission beyond 3G. In particular, the objective of the program has been to develop and explore a new type of radio system base ..."
Abstract
 Add to MetaCart
(Show Context)
Ericsson in Göteborg. Uppsala University acted as lead partner. The objectives were to develop and to study algorithms and system design for future wireless packet data transmission beyond 3G. In particular, the objective of the program has been to develop and explore a new type of radio system based on adaptive transmission. Wireless IP was initiated as a multiuniversity research project within the SSF PCC program in 2000. It became a SSF IT program funded by 10 MSEK from mid2002 to mid2005, after which the program received a 2.5year extension funded by 5 MSEK from mid 2006 to the end of 2008. During 20022003, the WIP program performed an intense design effort, which resulted in the design and evaluation of a downlink for an adaptive OFDM system. The work discussed and offered answers to the following questions: Can a balanced adaptive OFDMbased downlink be designed without too much overhead due to guardbands and control signaling? What multiuser scheduling gains may be attained in such systems? What spectral efficiency may be expected by using adaptive OFDMbased downlinks
DiversityEnhanced Equal Access – Considerable Throughput Gains With 1bit Feedback
"... Abstract — We investigate performance aspects of adaptive modulation and scheduling as the amount of channel feedback is reduced. We study throughput, fairness and the sensitivity to incorrect channel quantizations. A main finding is that the throughput of a cellular downlink using strict multiuser ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract — We investigate performance aspects of adaptive modulation and scheduling as the amount of channel feedback is reduced. We study throughput, fairness and the sensitivity to incorrect channel quantizations. A main finding is that the throughput of a cellular downlink using strict multiuser diversity does not degrade significantly when the channel information is heavily quantized. On the other hand, unfairness increases and due to an inherent sensitivity to incorrectly chosen quantization levels there is a risk of occasional drastic performance drops. Noting that fixedaccess schemes do not have the bad properties of multiuser diversity, but achieve unsatisfactory throughput, we propose a scheme combining the good aspects of multiuser diversity with the desirable properties of fixed access schemes. The result is a lowcomplexity scheduler and quantization policy that achieve large throughput gains as compared to fixed access without compromising fairness. I.