arXiv reader

Computation Efficiency Maximization in Wireless-Powered Mobile Edge Computing Networks

Fuhui Zhou and Rose Qingyang Hu

arXiv
2001.10234v2 · PDF · 11 February 2020
Categories
eess.SP, cs.IT, math.IT
Licence
CC0 1.0 The site shows the text.

Abstract

Energy-efficient computation is an inevitable trend for mobile edge computing (MEC) networks. Resource allocation strategies for maximizing the computation efficiency are critically important. In this paper, computation efficiency maximization problems are formulated in wireless-powered MEC networks under both partial and binary computation offloading modes. A practical non-linear energy harvesting model is considered. Both time division multiple access (TDMA) and non-orthogonal multiple access (NOMA) are considered and evaluated for offloading. The energy harvesting time, the local computing frequency, and the offloading time and power are jointly optimized to maximize the computation efficiency under the max-min fairness criterion. Two iterative algorithms and two alternative optimization algorithms are respectively proposed to address the non-convex problems formulated in this paper. Simulation results show that the proposed resource allocation schemes outperform the benchmark schemes in terms of user fairness. Moreover, a tradeoff is elucidated between the achievable computation efficiency and the total number of computed bits. Furthermore, simulation results demonstrate that the partial computation offloading mode outperforms the binary computation offloading mode and NOMA outperforms TDMA in terms of computation efficiency.

Items 19

Remark

Remark 1

W7ZD

The major application scenarios for wireless powered MEC networks include two cases. One is in the wireless sensor or wearable networks where the mobile sensors and wearable computing devices are with milliwatt power consumption while they need to perform computation tasks, such as environmental parameter or physical condition monitoring [S., Bi]. The other one is in the areas, such as wildernesses and complex terrains, where the government needs to keep monitoring the environment so that the corresponding strategies can be taken to protect the environment. In those areas, neither cable charging or battery replacement can be conveniently established nor the cost for establishing cable charging systems is affordable. The wireless powered MEC network becomes a desirable alternative [C., You].

Theorem

Theorem 1

ETVL

In wireless powered MEC networks with TDMA under the partial computation offloading mode, the maximum CE under the max-min fairness criterion is achieved when P_s=P_{th}.

Remark

Remark 2

J1WX

It can be seen from Theorem 1 that the maximum CE achieved under the max-min fairness criterion increases with the transmission power of the power station in the wireless powered MEC networks with TDMA. If the transmission power level of the power station is not large enough for achieving the maximum harvested power of the user, the CE can be increased by increasing the transmission power of the power station.

Lemma

Lemma 1

9PMX

The optimal solution of \mathbf{P}_{{1}} can be obtained if and only if the following equation holds.

\begin{align}\ &{\mathop {\max }\limits_{{\tau _0},{\tau _k},{P_k},{f_k}} }\ {\mathop {\min }\limits_{k \in{\cal K}}}\ R_k\left( {{\tau _k},{P_k},{f_k}} \right)-\eta^{*}E_k\left( {\tau _0, {\tau _k},{P_k},{f_k}} \right)\\ &= {\mathop {\min }\limits_{k \in{\cal K}}}\ R_k\left( {{\tau _k^{*}},{P_k^{*}},{f_k^{*}}} \right)-\eta^{*}E_k\left( {\tau _0^{*}, {\tau _k}^{*},{P_k}^{*},{f_k^{*}}} \right)=0, \end{align}
Lemma

Lemma 2

OXS0

\mathbf{P}_{{3}} is convex and can be efficiently solved by using the convex optimization tool [S., P., Boyd].

Theorem

Theorem 2

WWUN

In the wireless powered MEC systems with TDMA, the optimal local computation frequency f_k^{*} and the optimal offloading power P_k^{*} of the kth user for maximizing the CE under the max-min fairness criterion have the following mathematical expressions:

\begin{align}\ &f_k^{*} = \sqrt {\frac{{{\lambda _k} + {\theta _k}}}{{3C{\gamma _c}\left( {{\rho _k} + {\theta _k}\eta } \right)}}};\\ &P_k^{*} = \left\{ \begin{array}{l} \begin{array}{*{20}{c}} {0,}&{\text{if}\ {\tau _k} = 0} \end{array}\\ \begin{array}{*{20}{c}} {{{\left[ {\frac{{\left( {{\lambda _k} + {\theta _k}} \right)B}}{{\zeta{v_k}\ln 2\left( {{\rho _k} + {\theta _k}\eta } \right)}} - \frac{{\sigma _0^2}}{{{g_k}}}} \right]}^ + },}&{\text{otherwise}}; \end{array} \end{array} \right. \end{align}

where \lambda _k\geq0, \rho _k\geq0 and \theta _k\geq0 are the dual variables corresponding to the constraints given by \left(7\rm{b}\right), \left(7\rm{c}\right), and \left(7\rm{d}\right), respectively.

Used by Remark 3
Proof of Theorem 2

See Appendix A.

Remark

Remark 3

OEFV

It can be seen from Theorem 2 that the kth user chooses to offload its computation task only when the channel between the kth user and the MEC server is good enough, namely, g_k\geq {\left[ {\sigma _0^2\zeta{v_k}\ln 2\left( {{\rho _k} + {\theta _k}\eta } \right)} \right]}/{\left( {{\lambda _k} + {\theta _k}} \right)B}. Moreover, the local computation frequency decreases with the increase of \eta, which is related to the CE. It indicates that the users prefer to offload their computation task to the MEC server in order to improve the CE. Furthermore, it can be seen that the users prefer to offload their computation task when the local CPU frequency is too high, namely, f_k^{*} \ge \sqrt {\frac{{\sigma _0^2\zeta{v_k}\ln 2}}{{{g_k}3C{\gamma _c}B}}}. Additionally, when f_k^{*} < \sqrt {\frac{{\sigma _0^2\zeta{v_k}\ln 2}}{{{g_k}3C{\gamma _c}B}}}, the optimal offloading power P_k^{*} is zero. It means that the users only perform local computation in order to maximize their CE.

Theorem

Theorem 3

F5NN

For the given \lambda _k, \rho _k, \beta and \theta _k, in order to maximize the Lagrangian of \mathbf{P}_{{3}}, the optimal EH time \tau_0^{*} and computation offloading time \tau_k^{*} need to satisfy the following equations.

\begin{align}\ &\tau _0^{*} = \left\{ \begin{array}{l} \begin{array}{*{20}{c}} {0,}&{\text{if}\ z\left( {{\rho _k},\beta ,{\theta _k}} \right) < 0}, \end{array}\\ \begin{array}{*{20}{c}} { \in \left[ {0,T} \right),}&{\text{if} \ z\left( {{\rho _k},\beta ,{\theta _k}} \right) = 0}, \end{array} \end{array} \right. \\ &z\left( {{\rho _k},\beta ,{\theta _k}} \right) = \sum\limits_{k = 1}^K {{\rho _k}} \left( {{P_{E,k}} - {P_{r,k}}} \right) - \beta - \sum\limits_{k = 1}^K {{\theta _k}\eta } {P_{r,k}},\\ &{P_{E,k}} ={\frac{{P_k^{\max }}}{{\exp \left( { - \mu {P_0} + \psi } \right)}}\left( {\frac{{1 + \exp \left( { - \mu {P_0} + \psi } \right)}}{{1 + \exp \left( { - \mu {h_k}{P_{th}} + \psi } \right)}} - 1} \right)}, \end{align}
\begin{align}\ &\tau _k^{*} = \left\{ \begin{array}{l} \begin{array}{*{20}{c}} {{\rm Z} ,}&{\text{if} \ \frac{{\sigma _0^2\zeta{v_k}{\theta _k}\eta \ln 2}}{{\left( {{\lambda _k} + {\theta _k}} \right)B}} > \omega^{opt} }, \end{array}\\ \begin{array}{*{20}{c}} { \in \left[ {0,{\rm Z} } \right],}&{\text{if} \ \frac{{\sigma _0^2\zeta{v_k}{\theta _k}\eta \ln 2}}{{\left( {{\lambda _k} + {\theta _k}} \right)B}} = \omega^{opt} }, \end{array}\\ \begin{array}{*{20}{c}} {0,}&{\text{if} \ \frac{{\sigma _0^2\zeta{v_k}{\theta _k}\eta \ln 2}}{{\left( {{\lambda _k} + {\theta _k}} \right)B}} < \omega^{opt} }, \end{array} \end{array} \right. \\ &{\rm Z } = \frac{{{\Phi _k}\left( {\tau _0^{opt},{P_{th}}} \right) - \tau _0^{opt}{P_{r,k}} - T{\gamma _c}{{\left( {f_k^{opt}} \right)}^3}}}{{{P_{c,k}} + P_k^{opt}}}, \end{align}

where \omega^{*} is the solution of the following equation.

\begin{aligned}\ &\frac{{B\left( {{\lambda _k} + {\theta _k}} \right)}}{{{v_k}}}{\log _2}\left[ {\frac{{\left( {{\lambda _k} + {\theta _k}} \right)Bw}}{{\zeta{v_k}{\theta _k}\eta \sigma _0^2\ln 2}}} \right] - \frac{{\left( {{\lambda _k} + {\theta _k}} \right)B}}{{{v_k}\ln 2}}\\ &+ \frac{{\zeta\left( {{\rho _k} + {\theta _k}\eta } \right) \sigma _0^2}}{w} - \zeta\left( {{\rho _k} + {\theta _k}\eta } \right) {P_{c,k}} - \beta = 0. \end{aligned}
Proof of Theorem 3

See Appendix B.

Theorem

Theorem 4

FX13

In the wireless powered MEC systems with TDMA under the partial computation offloading mode, the maximum CE under the max-min fairness criterion is given as

\begin{align}\ &\Upsilon ^{*}= \left\{ \begin{array}{l} \begin{array}{*{20}{c}} {0,}&{\text{if}\ \sum\limits_{k = 1}^K { {\theta _k}} > 1,} \end{array}\\ \begin{array}{*{20}{c}} {{\Lambda ^*,}}&{\text{if}\ \sum\limits_{k = 1}^K { {\theta _k}} \le 1;} \end{array} \end{array} \right. \\ &\Lambda ^*={\mathop {\min }\limits_{k \in{\cal K}}}\ R_k\left( {{\tau _k^{*}},{P_k^{*}},{f_k^{*}}} \right)-\eta E_k\left( {\tau _0^{*}, {\tau _k}^{*},{P_k}^{*},{f_k^{*}}} \right), \end{align}
Proof of Theorem 4

Since \mathbf{P}_{{3}} is convex and the Slater’s conditions are satisfied, the Lagrangian of \mathbf{P}_{{3}} should be upper-bound with respect to \Upsilon. When \sum\limits_{k = 1}^K { {\theta _k}} > 1, the Lagrangian decreases with \Upsilon. Thus, the maximum of Lagrangian is achieved with \Upsilon=0 since \Upsilon\geq0. When \sum\limits_{k = 1}^K { {\theta _k}} \leq 1, the maximum of Lagrangian is obtained when the optimal solution is achieved.

Remark

Remark 4

M2D0

By using Theorem 1 and introducing auxiliary variables y_k, it is seen that \mathbf{P}_{{1}} is a generalized fractional programming problem [G., M., N.Guerekata]. Moreover, Algorithm 1 is proposed for solving \mathbf{P}_{{1}} based on the Dinkelbach’s method. Thus, according to [G., M., N.Guerekata], Algorithm 1 can converge when updating \eta _{}^{n}. The detail proof for the convergence can be seen in [G., M., N.Guerekata].

Lemma

Lemma 3

DB2U

In wireless powered MEC systems with TDMA under the binary offloading mode, the maximum CE under the max-min fairness criterion is achieved when P_s=P_{th}.

Theorem

Theorem 5

KRAB

In the wireless powered MEC systems with TDMA under the binary computation offloading mode, the optimal operational mode selection index has the following form

\begin{align}\ &{\alpha _k}^{*}= \left\{ \begin{array}{l} 0,\text{if}\ {F_{1,k}} < {F_{2,k}},\\ 1,\text{otherwise}; \end{array} \right. \\ &{F_{1,k}} = \left( {{\lambda _{{k}}} + {\chi _k}} \right)\frac{{B{\tau _k}}}{{{v_k}}}{\log _2}\left( {1 + \frac{{{g_k}{{{y}}_k}}}{{{\tau _k}\sigma _0^2}}} \right)\\ &\ \ \ \ \ \ \ - \zeta \left( {{\mu _k} + {\chi _k}\eta } \right)\left( {{{{y}}_k} + {\tau _k}{P_{c,k}}} \right) - \upsilon {\tau _k},\\ &{F_{2,k}} = \left( {{\lambda _{{k}}} + {\chi _k}} \right)\frac{{T{f_k}}}{C} - \left( {{\mu _k} + {\chi _k}\eta } \right)T{\gamma _c}f_k^3, \end{align}

where \lambda _{{k}}\geq0, \mu _k\geq0, \upsilon\geq0 and \chi _k\geq0 are the dual variables associated with the constraints given by \left(18\rm{b}\right), \left(18\rm{c}\right), \left(18\rm{d}\right) and \left(17\rm{d}\right).

Proof of Theorem 5

Theorem 5 can be readily proved from the derivations of the Lagrangian with respect to {\alpha _k}. The proof is omitted due to the space limit.

Remark

Remark 5

H3R7

The proof for the convergence of Algorithm 2 when updating \alpha_k can be obtained from two facts. One is that \mathbf{P}_{{4}} is a generalized fractional programming problem, which can be solved by using Algorithm 2 based on the Dinkelbach’s method. Algorithm 2 can converge when iteratively updating \eta [G., M., N.Guerekata]. This indicates that the objective function of \mathbf{P}_{{6}} is nondecreasing, which can be easily proved by using the property of the generalized fractional programming problem [G., M., N.Guerekata]. The other is that \mathbf{P}_{{6}} is convex in terms of \alpha_k. Thus, in each iteration, there is only an optimal solution of \alpha_k. Due to those two facts, it is easy to prove that Algorithm 2 is converged when updating \alpha_k.

Lemma

Lemma 4

YR0V

In wireless powered MEC systems with NOMA under the partial computation offloading mode, the maximum CE under the max-min fairness criterion is always achieved when P_s=P_{th}.

Theorem

Theorem 6

T9H4

In the wireless powered MEC systems with NOMA under the binary offloading mode, the optimal operational mode selection index has the form given by eq. (26) in the top of the next page,

The paper has a figure here, which this page does not show.

where \varpi _{{k}}\geq0, \lambda _{{k}}\geq0, \omega _k\geq0 and \mu _k\geq0 are the dual variables associated with the constraints given by \left(25\rm{b}\right) - \left(25\rm{g}\right), respectively.

Source and licence

The text on this page is from arXiv:2001.10234v2 by Fuhui Zhou and Rose Qingyang Hu, under the licence CC0 1.0. The corpus made these changes: it read the items out of the LaTeX source, it expanded the macros of the paper, it put each proof after its statement, and it converted the text to HTML. The formulas are the TeX of the paper, and KaTeX renders them. A formula or a reference can look different from the PDF. The page is published under CC BY 4.0.

The record of each item is in analysis/objects/2001/2001.10234.jsonl.