skip to main content

Title: Self-Organizing mm Wave Networks: A Power Allocation Scheme Based on Machine Learning
Millimeter-wave (mmWave) communication is anticipated to provide significant throughout gains in urban scenarios. To this end, network densification is a necessity to meet the high traffic volume generated by smart phones, tablets, and sensory devices while overcoming large pathloss and high blockages at mmWaves frequencies. These denser networks are created with users deploying small mm Wave base stations (BSs) in a plug-and-play fashion. Although, this deployment method provides the required density, the amorphous deployment of BSs needs distributed management. To address this difficulty, we propose a self-organizing method to allocate power to mm Wave BSs in an ultra dense network. The proposed method consists of two parts: clustering using fast local clustering and power allocation via Q-learning. The important features of the proposed method are its scalability and self-organizing capabilities, which are both important features of 5G. Our simulations demonstrate that the introduced method, provides required quality of service (QoS) for all the users independent of the size of the network.
Authors:
;
Award ID(s):
1642865
Publication Date:
NSF-PAR ID:
10076909
Journal Name:
2018 11th Global Symposium on Millimeter Waves (GSMM)
Page Range or eLocation-ID:
1 to 4
Sponsoring Org:
National Science Foundation
More Like this
  1. We consider intermittently connected vehicular networks (ICVNs) in which base stations (BSs) are installed along the highway to connect moving vehicles with internet. Due to the deployment cost, it is hard to cover the entire highway with BSs. To minimize the outage time in the uncovered area (UA), several cooperative store-carry-forward (CSCF) schemes have been proposed in which a vehicle is selected to act as a relay by buffering data to be relayed to a target vehicle in the UA. In this paper, we propose an energy-efficient full-duplex (FD) CSCF scheme that exploits the relay ability to receive and transmit simultaneously to improve the effective communication time, Te, between the relay and the target vehicle. Accordingly, it can minimize the outage time and deliver more data to the the target vehicle. In addition, the power allocation that minimizes the transmission cost (TC) under the required rates constraints is found. The problem is formulated as a geometric program (GP) and globally solved using the interior-point method. As compared to the half-duplex CSCF scheme, simulation results show that the proposed FD scheme offers more effective time, more successfully delivered data in the UA and lower TC.
  2. The sensitivity to blockages at millimeter-wave (mm-wave) frequencies is very different from that at sub-6 GHz frequencies. The blockages affect the user-to-base station (BS) associations and the resulting association regions of the BSs in the network. This in turn alters the load, i.e., the total number of users associated to a BS. In this paper, we use a stochastic blockage model to analyze such effects. We characterize the variation in the load as a function of the blockage environment in a stochastic geometric setting. Our analysis indicates that in the extreme cases of total blocking and no blocking, the mean load on the tagged mm-wave BS is identical to that of a sub-6 GHz BS for a given BS and user density. For intermediate blockage environments, the mean load on the tagged mm-wave BS is found to be less than that on a sub-6 GHz BS. Using Monte-Carlo simulations, we establish that the existing analytical models for load characterization in mm-wave networks result in overestimation of the load per BS and underestimation of the achievable rate.
  3. To provide a reliable wireless uplink for users in a given ground area, one can deploy Unmanned Aerial Vehicles (UAVs) as base stations (BSs). In another application, one can use UAVs to collect data from sensors on the ground. For a power efficient and scalable deployment of such flying BSs, directional antennas can be utilized to efficiently cover arbitrary 2-D ground areas. We consider a large-scale wireless path-loss model with a realistic angle-dependent radiation pattern for the directional antennas. Based on such a model, we determine the optimal 3-D deployment of N UAVs to minimize the average transmit-power consumption of the users in a given target area. The users are assumed to have identical transmitters with ideal omnidirectional antennas and the UAVs have identical directional antennas with given half-power beamwidth (HPBW) and symmetric radiation pattern along the vertical axis. For uniformly distributed ground users, we show that the UAVs have to share a common flight height in an optimal power-efficient deployment, by simulations. We also derive in closed-form the asymptotic optimal common flight height of N UAVs in terms of the area size, data-rate, bandwidth, HPBW, and path-loss exponent.
  4. A self-organizing map (SOM) is a type of competitive artificial neural network, which projects the high- dimensional input space of the training samples into a low-dimensional space with the topology relations preserved. This makes SOMs supportive of organizing and visualizing complex data sets and have been pervasively used among numerous disciplines with different applications. Notwithstanding its wide applications, the self-organizing map is perplexed by its inherent randomness, which produces dissimilar SOM patterns even when being trained on identical training samples with the same parameters every time, and thus causes usability concerns for other domain practitioners and precludes more potential users from exploring SOM based applications in a broader spectrum. Motivated by this practical concern, we propose a deterministic approach as a supplement to the standard self-organizing map. In accordance with the theoretical design, the experimental results with satellite cloud data demonstrate the effective and efficient organization as well as simplification capabilities of the proposed approach.
  5. Recent years have witnessed a growing body of research on autonomous activity recognition models for use in deployment of mobile systems in new settings such as when a wearable system is adopted by a new user. Current research, however, lacks comprehensive frameworks for transfer learning. Specifically, it lacks the ability to deal with partially available data in new settings. To address these limitations, we propose {\it OptiMapper}, a novel uninformed cross-subject transfer learning framework for activity recognition. OptiMapper is a combinatorial optimization framework that extracts abstract knowledge across subjects and utilizes this knowledge for developing a personalized and accurate activity recognition model in new subjects. To this end, a novel community-detection-based clustering of unlabeled data is proposed that uses the target user data to construct a network of unannotated sensor observations. The clusters of these target observations are then mapped onto the source clusters using a complete bipartite graph model. In the next step, the mapped labels are conditionally fused with the prediction of a base learner to create a personalized and labeled training dataset for the target user. We present two instantiations of OptiMapper. The first instantiation, which is applicable for transfer learning across domains with identical activity labels,more »performs a one-to-one bipartite mapping between clusters of the source and target users. The second instantiation performs optimal many-to-one mapping between the source clusters and those of the target. The many-to-one mapping allows us to find an optimal mapping even when the target dataset does not contain sufficient instances of all activity classes. We show that this type of cross-domain mapping can be formulated as a transportation problem and solved optimally. We evaluate our transfer learning techniques on several activity recognition datasets. Our results show that the proposed community detection approach can achieve, on average, 69%$ utilization of the datasets for clustering with an overall clustering accuracy of 87.5%. Our results also suggest that the proposed transfer learning algorithms can achieve up to 22.5% improvement in the activity recognition accuracy, compared to the state-of-the-art techniques. The experimental results also demonstrate high and sustained performance even in presence of partial data.« less