| 1. |
- Abrahamsson, Olle, et al.
(författare)
-
Opinion Dynamics with Random Actions and a Stubborn Agent
- 2019
-
Ingår i: CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS. - : IEEE. - 9781728143002 ; , s. 1486-1490
-
Konferensbidrag (refereegranskat)abstract
- We study opinion dynamics in a social network with stubborn agents who influence their neighbors but who themselves always stick to their initial opinion. We consider first the well-known DeGroot model. While it is known in the literature that this model can lead to consensus even in the presence of a stubborn agent, we show that the same result holds under weaker assumptions than has been previously reported. We then consider a recent extension of the DeGroot model in which the opinion of each agent is a random Bernoulli distributed variable, and by leveraging on the first result we establish that this model also leads to consensus, in the sense of convergence in probability, in the presence of a stubborn agent. Moreover, all agents opinions converge to that of the stubborn agent.
|
|
| 2. |
- Alegria, Juan Vidal, et al.
(författare)
-
Cramér-Rao Lower Bounds for Positioning with Large Intelligent Surfaces using Quantized Amplitude and Phase
- 2020
-
Ingår i: 2019 53rd Asilomar Conference on Signals, Systems, and Computers. - 9781728143002 - 9781728143019 ; , s. 10-14
-
Konferensbidrag (refereegranskat)abstract
- We envision the use of large intelligent surface (LIS) technology, which is a promising concept that goes beyond massive multiple-input multiple-output (MIMO), for positioning applications due to its ability to focus energy in the 3D space. The Cramér-Rao lower bounds (CBLBs) for positioning using a LIS which can resolve amplitude and phase with full resolution have already been determined in the previous literature. However, in real applications, and specially if we consider cheap hardware components to enable the deployment of LIS at reasonable costs, the phase and amplitude have to be quantized before any information can be extracted from them. Furthermore, the phase information is more difficult to resolve due to phase noise, non-coherence, etc. In this paper we compute the CRLBs for positioning using LIS with quantized phase and amplitude. We also derive analytical bounds for the CRLB for positioning with LIS when all phase information is disregarded and amplitude is measured with full resolution. We present numerical results in the form of tables including the CRLB loss due to the different quantization resolutions, which can serve as a design guideline for hardware developers.
|
|
| 3. |
- Björnson, Emil, Professor, 1983-, et al.
(författare)
-
Utility-Based Precoding Optimization Framework for Large Intelligent Surfaces
- 2019
-
Ingår i: CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS. - : IEEE. - 9781728143002 ; , s. 863-867
-
Konferensbidrag (refereegranskat)abstract
- The spectral efficiency of wireless networks can be made nearly infinitely large by deploying many antennas, but the deployment of very many antennas requires new topologies beyond the compact and discrete antenna arrays used by conventional base stations. In this paper, we consider the large intelligent surface scenario where small antennas are deployed on a large and dense two-dimensional grid. Building on the heritage of MIMO, we first analyze the beamwidth and sidelobes when transmitting from large intelligent surfaces. We compare different precoding schemes and determine how to optimize the transmit power with respect to different utility functions.
|
|
| 4. |
- Chakraborty, Sucharita, et al.
(författare)
-
Centralized and Distributed Power Allocation for Max-Min Fairness in Cell-Free Massive MIMO
- 2019
-
Ingår i: CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS. - : IEEE. - 9781728143002 ; , s. 576-580
-
Konferensbidrag (refereegranskat)abstract
- Cell-free Massive MIMO systems consist of a large number of geographically distributed access points (APs) that serve users by coherent joint transmission. Downlink power allocation is important in these systems, to determine which APs should transmit to which users and with what power. If the system is implemented correctly, it can deliver a more uniform user performance than conventional cellular networks. To this end, previous works have shown how to perform system-wide max-min fairness power allocation when using maximum ratio precoding. In this paper, we first generalize this method to arbitrary precoding, and then train a neural network to perform approximately the same power allocation but with reduced computational complexity. Finally, we train one neural network per AP to mimic system-wide max-min fairness power allocation, but using only local information. By learning the structure of the local propagation environment, this method outperforms the state-of-the-art distributed power allocation method from the Cell-free Massive MIMO literature.
|
|
| 5. |
- Chakraborty, Sucharita, et al.
(författare)
-
Centralized and Distributed Power Allocation for Max-Min Fairness in Cell-Free Massive MIMO
- 2019
-
Ingår i: CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS. - : IEEE. - 9781728143002 ; , s. 576-580
-
Konferensbidrag (refereegranskat)abstract
- Cell-free Massive MIMO systems consist of a large number of geographically distributed access points (APs) that serve users by coherent joint transmission. Downlink power allocation is important in these systems, to determine which APs should transmit to which users and with what power. If the system is implemented correctly, it can deliver a more uniform user performance than conventional cellular networks. To this end, previous works have shown how to perform system-wide max-min fairness power allocation when using maximum ratio precoding. In this paper, we first generalize this method to arbitrary precoding, and then train a neural network to perform approximately the same power allocation but with reduced computational complexity. Finally, we train one neural network per AP to mimic system-wide max-min fairness power allocation, but using only local information. By learning the structure of the local propagation environment, this method outperforms the state-of-the-art distributed power allocation method from the Cell-free Massive MIMO literature.
|
|
| 6. |
- Chowdhury, Sohini Roy, et al.
(författare)
-
Automated Augmentation with Reinforcement Learning and GANs for Robust Identification of Traffic Signs using Front Camera Images
- 2019
-
Ingår i: Conference Record - Asilomar Conference on Signals, Systems and Computers. - 1058-6393. - 9781728143002 ; 2019-November, s. 79-83
-
Konferensbidrag (refereegranskat)abstract
- Traffic sign identification using camera images from vehicles plays a critical role in autonomous driving and path planning. However, the front camera images can be distorted due to blurriness, lighting variations and vandalism which can lead to degradation of detection performances. As a solution, machine learning models must be trained with data from multiple domains, and collecting and labeling more data in each new domain is time consuming and expensive. In this work, we present an end-to-end framework to augment traffic sign training data using optimal reinforcement learning policies and a variety of Generative Adversarial Network (GAN) models, that can then be used to train traffic sign detector modules. Our automated augmenter enables learning from transformed nightime, poor lighting, and varying degrees of occlusions using the LISA Traffic Sign and BDD-Nexar dataset. The proposed method enables mapping training data from one domain to another, thereby improving traffic sign detection precision/recall from 0.70/0.66 to 0.83/0.71 for nighttime images.
|
|
| 7. |
- Ettefagh, Yasaman, 1989, et al.
(författare)
-
All-Digital Massive MIMO Uplink and Downlink Rates under a Fronthaul Constraint
- 2019
-
Ingår i: Conference Record - Asilomar Conference on Signals, Systems and Computers. - 1058-6393. - 9781728143002 ; 2019-November, s. 416-420
-
Konferensbidrag (refereegranskat)abstract
- We characterize the rate achievable in a bidirectional quasi-static link where several user equipments communicate with a massive multiple-input multiple-output base station (BS). In the considered setup, the BS operates in full-digital mode, the physical size of the antenna array is limited, and there exists a rate constraint on the fronthaul interface connecting the (possibly remote) radio head to the digital baseband processing unit. Our analysis enables us to determine the optimal resolution of the analog-todigital and digital-to-analog converters as well as the optimal number of active antenna elements to be used in order to maximize the transmission rate on the bidirectional link, for a given constraint on the outage probability and on the fronthaul rate. We investigate both the case in which perfect channel-state information is available, and the case in which channel-state information is acquired through pilot transmission, and is, hence, imperfect. For the second case, we present a novel rate expression that relies on the generalized mutual-information framework.
|
|
| 8. |
- Hatami, Mohammad, et al.
(författare)
-
Online Caching Policy with User Preferences and Time-Dependent Requests: A Reinforcement Learning Approach
- 2019
-
Ingår i: CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS. - : IEEE. - 9781728143002 ; , s. 1384-1388
-
Konferensbidrag (refereegranskat)abstract
- Content caching is a promising approach to reduce data traffic in the back-haul links. We consider a system where multiple users request items from a cache-enabled base station that is connected to a cloud. The users request items according to the user preferences in a time-dependent fashion, i.e., a user is likely to request the next chunk (item) of the file requested at a previous time slot. Whenever the requested item is not in the cache, the base station downloads it from the cloud and forwards it to the user. In the meanwhile, the base station decides whether to replace one item in the cache by the fetched item, or to discard it. We model the problem as a Markov decision process (MDP) and propose a novel state space that takes advantage of the dynamics of the users requests. We use reinforcement learning and propose a Q-learning algorithm to find an optimal cache replacement policy that maximizes the cache hit ratio without knowing the popularity profile distribution, probability distribution of items, and user preference model. Simulation results show that the proposed algorithm improves the cache hit ratio compared to other baseline policies.
|
|
| 9. |
- Liva, Gianluigi, et al.
(författare)
-
Coded Slotted Aloha over the On-Off Fading Channel: Performance Bounds
- 2019
-
Ingår i: Conference Record - Asilomar Conference on Signals, Systems and Computers. - 1058-6393. - 9781728143002 ; 2019-November, s. 31-35
-
Konferensbidrag (refereegranskat)abstract
- The on-off fading channel is used as a proxy to gain insights on the performance of coded slotted Aloha (CSA) access protocols on the block flat fading channel. Bounds on the performance of CSA are developed, which include both the on-off fading effect and multi-packet reception capabilities at the receiver. The bounds apply to the case where a non-vanishing packet loss probability can be tolerated. A comparison of simulation results for actual CSA schemes with the developed bounds is provided.
|
|
| 10. |
- Mazloum, Nafiseh, et al.
(författare)
-
Taking Cellular IoT Energy Efficiency to the Next Level
- 2020
-
Ingår i: 2019 53rd Asilomar Conference on Signals, Systems, and Computers. - 9781728143002 - 9781728143019 ; , s. 1951-1956
-
Konferensbidrag (refereegranskat)abstract
- We present and analyze a new approach to improve energy efficiency in cellular IoT devices. The new approach consists of employment of ultra low power wake-up receivers in combination with robust non-coherent modulation. The results show that we can achieve very competitive power consumption, reducing it by up to 50 times, compared to two conventional power saving solutions we use as references. These large power savings are particularly important when the device needs to be highly available, in terms of wake-up time, and has a tight delay requirement. In these situations, the large power savings can be traded for shorter duty-cycle lengths. This translates to correspondingly shorter wake-up delays and our results show that they can be reduced well over 100 times, at the same average power consumption.
|
|
