Error with find Function in MATLAB: "Incorrect number or types of inputs or outputs"

Hello,

I am encountering an error in MATLAB when using the find function in my code. The error message is:

Error using find Incorrect number or types of inputs or outputs for function 'find'.

The error occurs in the following function:

load(MyData.m)
function SNR_matrix = Compute_SNR(N, numNodes, Devices,A,zeta_m_n,h0, distances,alpha,RIS_elements,arg1,arg2,Theta_node,P_v,Noise_up)
 % SNR_matrix = zeros(N, numNodes, Devices);
SNR_matrix = optimexpr(N, numNodes, Devices);  

for v = 1:Devices
    for sb = 1:N
        for node = 1:numNodes

            % Find the assigned node for subtask sb of vehicle v
            assigned_node = find(A(sb, :, v), 1);

            % Find the covering UAV for the vehicle
            covering_uav = find(zeta_m_n(v, :), 1);

            if assigned_node == covering_uav
                % Direct communication
                   SNR =  sqrt(h0 * distances(v, covering_uav)^(-alpha));
                   SNR_matrix(sb, assigned_node, v) = (P_v*norm(SNR)^2)/Noise_up;
            else
                % Indirect communication through RIS
                % Extract channel gain from vehicle to covering UAV
                channel_gains_vehicle_to_covering = zeros(RIS_elements, 1);
                for k = 1:RIS_elements
                    channel_gains_vehicle_to_covering(k) = arg1(v, covering_uav, k);
                end

                % Extract channel gain from covering UAV to assigned node
                arg2_covering_to_assigned = zeros(RIS_elements, 1);
                for k = 1:RIS_elements
                    arg2_covering_to_assigned(k) = arg2(covering_uav, assigned_node, k);
                end
                h_m_to_n_prime_RIS_phase_shifted = Theta_node{covering_uav} * channel_gains_vehicle_to_covering;
                effective_channel = h_m_to_n_prime_RIS_phase_shifted .* arg2_covering_to_assigned;

             
                effective_power = norm(effective_channel)^2;

                % Calculate SNR
                SNR = P_v * effective_power / Noise_up;
                SNR_matrix(sb, assigned_node, v) = SNR;


            end
        end
    end
end

end

and ''A'' is my optimization variable

A = optimvar('A', N, NumNodes, Devices, 'Type', 'integer', 'LowerBound', 0, 'UpperBound', 1);

Unrecognized function or variable 'N'.

How can I resolve this error? I have used fcn2optimexpr but the SNR expression becomes non-linear.

Is there a solution to avoid using find and keep the SNR linear?

Any Help will be appreciated.

Thanks in advance!

[Torsten]

How do you use "Compute_SNR" in your code ?

Somehow like

SNR_matrix = fcn2optimexpr(@Compute_SNR,K, N, M, A, z,...)

after defining all the inputs for the function ?

[Maria]

Thank you for responding me.

Just i used a simple call

SNR_matrix = Compute_SNR(K, N, M, A, z,...);

[Torsten]

That's not possible. You must define a function that can be evaluated using optimization variables - thus you must define the SNR_matrix as output of a fcn2optimexpr.

[Maria]

Thank u again.

Yes this time i didn't get the same error, should i apply it for each function that contains my optimization variable A??

[Torsten]

If you have a function with optimization variables as input and there are operations/functions within this function that cannot be applied to optimization variables (like "find" in the case posted), you have to use "fcn2optimexpr". I think @Walter Roberson gave you a link where the possible operations on optimization variables (without using "fcn2optimexpr") are listed.

[Maria]

yes, i understand now. Thank you very much.

[Maria]

Hello @Torsten

Could you please help me to solve this issue.

I need to linearize some constraints to use the linprog solver. The problem involves constraints with max functions and products of variables, which introduce nonlinearity.

Here is a portion of my code where I define the variables and constraints:

T_wait = optimvar('T_wait', K, N, M);
T_cmpl = optimvar('T_cmpl', K, N, M);

% Auxiliary variable for max function
z = optimvar('z', K, N, M);

% Define constraints
for m = 1:M
    for n = 1:K
        for k = 1:N
            % Non-negativity constraint
            prob.Constraints.(['constr_wait_nonneg_' num2str(m) '_' num2str(n) '_' num2str(k)]) = T_wait(n, k, m) >= 0;

            % Constraints for max function
            prob.Constraints.(['constr_wait_max_' num2str(m) '_' num2str(n) '_' num2str(k)]) = T_wait(n, k, m) >= z(n, k, m);
            prob.Constraints.(['constr_wait_zero_' num2str(m) '_' num2str(n) '_' num2str(k)]) = z(n, k, m) >= 0;

          % Constraints for nested max
                for k_prime = 1:N
                    if k ~= k_prime
                        % Ensure consistent dimensions
                        prob.Constraints.(['constr_wait_inner_' num2str(m) '_' num2str(n) '_' num2str(k) '_' num2str(k_prime)]) = ...
                            z(n, k, m) >= o(k_prime, k, m) * T_cmpl(n, k_prime, m) - uplink_time(n, k, m);
                    end
                end

            % Constraint: T_cmpl = T_end + T_wait
            prob.Constraints.(['constr_cmpl_' num2str(m) '_' num2str(n) '_' num2str(k)]) = ...
                T_cmpl(n, k, m) == End_time(n, k, m) + T_wait(n, k, m);
        end
    end
end

this is the error

Error using optim.problemdef.OptimizationProblem/solve
Unable to perform assignment because the left and right sides have a different number of elements.

Error in Solve_Linear_Problem (line 253)
[sol, fval, exitflag, output] = solve(prob);

Caused by:
    Failure in initial user-supplied nonlinear constraint function evaluation

[Torsten]

I doubt that the error message stems from the part of the code you posted.

And I wonder why you make life difficult for yourself: you relate constraint (m,n,k) with array index (n,k,m). Doesn't it confuse you ?

[Maria]

A already adjust it. Thank you for your response.

Answer 1

[Walter Roberson]

find() is not a supported operation on optimization variables.

https://www.mathworks.com/help/optim/ug/supported-operations-on-optimization-variables-expressions.html